Skip to main content
Home / Resources / Blog / Why Using In Silico Modeling for Cardiac Safety is the Next Big Thing

Why Using In Silico Modeling for Cardiac Safety is the Next Big Thing

The sky high cost of drug development means that late stage drug attrition is especially devastating. Thus, the ability to assess potential drug safety issues early— using in vitro data— would be very beneficial. The current paradigm for cardiac safety pharmacology is rooted in the preclinical ICH S7B and clinical ICH E14 guidelines. There have not been any withdrawal of drugs for causing the potentially life-threatening ventricular arrhythmia, “torsades de pointes” (TdP) since the pharmaceutical industry has adopted these guidelines.

Measuring drug-induced alterations to the electrocardiogram (ECG) QT interval is central to current guidelines. QT interval prolongation can result from pharmacological inhibition of the hERG ion channel. By being “hERGcentric” (or some might even say “hERGphobic”), this approach is overly conservative and can result in false positives. Moreover, the clinical approach to evaluating drug-induced cardiotoxicity— the thorough QT (TQT) study— is expensive and has a low positive predictive value. In this blog post, I’ll discuss the coming shift from using TQT studies towards leveraging early drug discovery data for in silico models to predict cardiotoxicity.

Why the globally used QT interval is an imperfect biomarker for assessing a drug’s proarrhythmic propensity

Inhibition of the delayed rectifier potassium current mediated by the hERG channel is positively correlated with clinically observable cardiac arrhythmias. But, there’s a lot more to drug-induced cardiotoxicity than just the hERG channel. Thus, there are drugs that potently inhibit hERG that do not present a TdP risk and drugs that do not inhibit hERG that do cause QT prolongation. The ideal approach to assessing drug-induced cardiotoxicity risk will have high specificity and sensitivity for drugs that truly pose a danger to patients. In addition, in the impending era of “Personalized Medicine,” the new paradigm would predict variability in the risk of cardiotoxicity for individuals that could be missed in a clinical trial.

Intrinsic and extrinsic factors for drug-induced cardiotoxicity

Comprehensive safety assessment of drugs will require examining multiple intrinsic and extrinsic drivers of cardiac biorhythms. The creation of action potentials in cardiac cells requires complex coordination of multiple currents that are mediated by sodium, calcium, and potassium ion channels. Drug-mediated alteration of any of these currents can either be pro- or anti-arrhythmic. A study that examined the effect of previously characterized drugs on multiple ion channels had a fewer false positives and false negatives compared to using the hERG assay alone.

In addition to examining intrinsic risk factors, it is also important to consider extrinsic factors related to concurrent usage of other medications, demographics, and physiological influences. Population variability— in terms of age, gender, genetics, ethnicity, renal/hepatic function, and electrolyte homeostasis— can dramatically impact how susceptible an individual is to developing drug-induced cardiotoxicity. In examining cases of drug-mediated torsades de pointes, extrinsic factors frequently contributed to the development of the arrhythmia. For example, one case study of a women who experienced repeated episodes of TdP implicated hypokalemia (low potassium levels) due to repeated vomiting and poor nutrition as a contributing factor.

Finally, drug-drug interactions (DDI) can play a significant role in drugs’ proarrhythmic effects due to changes in pharmacokinetics. One of the best known examples of QT prolongation due to a DDI is the combination of terfenadine and ketoconazole. The latter drug is a potent CYP3A4 inhibitor. Ketaconazole-mediated CYP3A4 inhibition causes a substantial increase in the concentration of terfenadine that can result in significant ECG repolarization abnormalities.

The paradigm shift towards in silico methods for cardiac safety testing

Fully assessing a drug’s potential for cardiotoxicity requires considering numerous intrinsic and extrinsic factors. Testing all potentially relevant scenarios through clinical trials is simply not possible. Biosimulation approaches including PBPK modeling and simulation can help test these scenarios in virtual clinical trials.

The Cardiac Safety Simulator: A comprehensive tool for risk assessment of drugs

The Cardiac Safety Simulator (CSS) uses drug-triggered cardiac ion-current disruption data, together with predicted in vivo exposure information to evaluate the factors influencing potential cardiac risk. It determines the drug candidate’s pro-arrhythmic potency by assessing its inhibition of multiple cardiac ion channels. CSS also factors in population variability. In addition, it assesses the influence of multiple drugs on ventricular ion current and simulated ECGs to account for participants who may be receiving treatment for more than one condition.

For more information, please read my AAPS Journal paper, “Early Drug Discovery Prediction of Proarrhythmia Potential and Its Covariates.” You might also be interested in my paper with my colleague, Dr. Amin Rostami, CPT: Pharmacometrics & Systems Pharmacology article “Interaction Between Domperidone and Ketoconazole: Toward Prediction of Consequent QTc Prolongation Using Purely In Vitro Information.” There is also an interesting commentary on this paper.

About the author

By: Sebastian Polak